File size: 2,315 Bytes
b3652fb cd88622 02cc722 81def50 02cc722 bef39ed 81def50 02cc722 c046173 81def50 c046173 8ff8f64 cd88622 b3652fb 81def50 b3652fb 81def50 903cef3 c046173 a1a4c08 c046173 a1a4c08 c046173 81def50 cd88622 c046173 2d5e944 c046173 81def50 cd88622 c046173 cd88622 903cef3 cd88622 b3652fb 81def50 b3652fb 81def50 b3652fb 81def50 b3652fb 81def50 f6e7520 81def50 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 |
import io
import time
import numpy as np
import cv2
import torch
import torchvision
from fastapi import FastAPI, File, UploadFile
from PIL import Image
import uvicorn
app = FastAPI()
# 🟢 Tải mô hình MobileNetDepth (MobileNet v3 Large)
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
model = torchvision.models.mobilenet_v3_large(pretrained=True).to(device)
model.eval()
@app.post("/analyze_path/")
async def analyze_path(file: UploadFile = File(...)):
# 🟢 Đọc file ảnh từ ESP32
image_bytes = await file.read()
image = Image.open(io.BytesIO(image_bytes)).convert("RGB")
# 🟢 Chuyển đổi ảnh thành tensor phù hợp với MobileNetDepth
transform = torchvision.transforms.Compose([
torchvision.transforms.Resize((224, 224)), # MobileNetDepth yêu cầu ảnh 224x224
torchvision.transforms.ToTensor(),
])
img_tensor = transform(image).unsqueeze(0).to(device)
# 🟢 Bắt đầu đo thời gian dự đoán Depth Map
start_time = time.time()
# 🟢 Dự đoán Depth Map với MobileNetDepth
with torch.no_grad():
depth_map = model(img_tensor)
depth_map = depth_map.squeeze().cpu().numpy()
end_time = time.time()
print(f"⏳ MobileNetDepth xử lý trong {end_time - start_time:.4f} giây")
# 🟢 Đo thời gian xử lý đường đi
start_detect_time = time.time()
command = detect_path(depth_map)
end_detect_time = time.time()
print(f"⏳ detect_path() xử lý trong {end_detect_time - start_detect_time:.4f} giây")
return {"command": command}
def detect_path(depth_map):
"""Phân tích đường đi từ ảnh Depth Map"""
h, w = depth_map.shape
center_x = w // 2
scan_y = h - 20 # Quét dòng gần đáy ảnh
left_region = np.mean(depth_map[scan_y, :center_x])
right_region = np.mean(depth_map[scan_y, center_x:])
center_region = np.mean(depth_map[scan_y, center_x - 20:center_x + 20])
if center_region > 200:
return "forward"
elif left_region > right_region:
return "left"
elif right_region > left_region:
return "right"
else:
return "backward"
# 🟢 Chạy server FastAPI
if __name__ == "__main__":
uvicorn.run(app, host="0.0.0.0", port=7860)
|